Comparison of BALF concentrations of ENA-78 and IP10 in patients with idiopathic pulmonary fibrosis and nonspecific interstitial pneumonia

Transcription

1 Respiratory Medicine (25) 99, Comparison of BALF concentrations of ENA-78 and IP1 in patients with idiopathic pulmonary fibrosis and nonspecific interstitial pneumonia Seiko Nakayama a, Hiroshi Mukae a,, Hiroshi Ishii a, Tomoyuki Kakugawa a, Kanako Sugiyama a, Noriho Sakamoto a, Takeshi Fujii a, Jun-ichi Kadota b, Shigeru Kohno a a Second Department of Internal Medicine, Nagasaki University School of Medicine, Nagasaki b Second Department of Internal Medicine, Oita University Faculty of Medicine, Oita, Japan Received 18 October 24 KEYWORDS Idiopathic pulmonary fibrosis; Nonspecific interstitial pneumonia; ENA-78; IP1 Summary Background: Epithelial neutrophil-activating peptide 78 (ENA-78) and interferon g- inducible protein 1 (IP1) belong to the CXC chemokine family and are considered to be important factors in idiopathic pulmonary fibrosis (). Idiopathic nonspecific interstitial pneumonia () and are the two largest subsets of idiopathic interstitial pneumonias (IIP). In patients with, the prognosis is generally good compared with. Therefore, the pathogenesis of seems to be different from that of, but this remains unclear. The aim of the present study was to evaluate the contribution of ENA-78 and IP1 in the two diseases. Methods: We measured the levels of ENA-78 and IP1 in serum and bronchoalveolar lavage fluid (BALF) of patients with (n ¼ 17), idiopathic (n ¼ 1) and healthy subjects (n ¼ 12) by enzyme-linked immunosorbent assays. Results: The level of ENA-78 in BALF was significantly higher in patients than in patients and controls. Serum levels of ENA-78 and BALF levels of IP1 in patients were significantly higher than in patients with and controls. In BALF of patients with, IP1 level significantly correlated with the absolute number of lymphocytes. In patients, BALF IP1 levels also correlated with the proportion of lymphocytes in BALF. Conclusion: Our results show distinct profiles of CXC chemokines in and, and suggest that these chemokines play an important role in inflammatory cell recruitment into the lung in patients with IIP. & 25 Elsevier Ltd. All rights reserved. Corresponding author. Tel.: ; fax: address: hmukae@net.nagasaki-u.ac.jp (H. Mukae) /$ - see front matter & 25 Elsevier Ltd. All rights reserved. doi:1.116/j.rmed

2 1146 S. Nakayama et al. Introduction The classification of idiopathic interstitial pneumonia (IIP) includes seven clinico-radiologic-pathological entities. Idiopathic pulmonary fibrosis () and nonspecific interstitial pneumonia () are the two largest subsets of IIP. 1 3 is distinguished from by the temporal uniformity of interstitial inflammation and/or fibrosis on histology. 1 3 The distinction between and is important for clinical decision-making because the prognosis is generally good and the response to corticosteroids and immunosuppressants is also good in patients with compared with. 1 3 It is well known that is associated with relative lymphocytosis with a predominance of CD8+ cells in bronchoalveolar lavage fluid (BALF) compared with. 4 6 Previous studies demonstrated that lymphocytosis correlated with the levels of interleukin (IL)-6 in BALF of patients with. 5 These results suggest that the pathogenesis of is different from that of, though their mechanisms remain elusive. Chemokines are important for leukocyte recruitment to the inflamed tissue. Chemokines are divided into four subfamilies depending on the position of the conserved cysteine residues (CXC, CC, C and CX 3 C). 7 It is well known that the CC chemokines such as monocyte chemoattractant protein 1 (MCP-1), macrophage inflammatory protein-1alpha (MIP-1a), and regulated upon activation, normal T cell expressed and secreted (RANTES), are closely related to the expression of adhesion molecules and the migration of inflammatory cells into the lung. There is sufficient evidence for the importance of these chemokines in These CC chemokines are also important in inflammatory cell recruitment to the lung in idiopathic. 13 The CXC chemokines such as epithelial neutrophilactivating peptide (ENA)-78, interferon g-inducible protein (IP)1 and IL-8 are also important factors in. 14,15 However, to our knowledge, there are no reports demonstrating the role of the CXC chemokine in idiopathic. We hypothesized that these CXC chemokines mediate the recruitment of inflammatory cells into the lung in patients with as well, similar to the behavior of the CC chemokine. In the present study, we measured the concentrations of ENA-78 and IP-1 in BALF and serum samples obtained from patients with and idiopathic to evaluate the contribution of these chemokines, and to determine whether there is a distinct profile of these factors in these diseases. Materials and methods Study population The 39 subjects of this study were patients and healthy enrolled in the Hospitals of Nagasaki University School of Medicine. They included 17 patients with idiopathic (14 males and 3 females, consisting of 5 current smokers, 3 ex-smokers and 9 nonsmokers, age; 61 1 years, mean7sd;), 1 with idiopathic (5 males and 5 females, consisting of one current smoker, 2 exsmokers and 7 nonsmokers, age; years), and 12 healthy (8 males and 4 females, 12 nonsmokers, age; 24 5 years). None of the enrolled patients had received steroid and other immunosuppressive therapy at the time of clinical sample collection. Patients with cancer in any organ and those suspected to have malignancy were excluded from the study. The diagnosis was pathologically confirmed using surgical lung biopsy specimens obtained in at least two different sites in all patients. The patients with usual interstitial pneumonia (UIP) and associated with collagen vascular diseases were excluded in this study. In all of 1 cases, the diagnosis was pathologically confirmed as fibrotic, consisting of 5 Group II and 5 Group III. 16 In these patients, the following high-resolution computed tomography (HRCT) criteria were used: (1) predominantly basal/subpleural distribution, (2) a mixture of reticular and ground-glass abnormalities, with traction bronchiectasis when ground-glass attenuation was prominent, (3) the absence of consolidation or nodules. All patients with fulfilled these criteria. There were no significant differences in the mean %VC, FEV 1 %, %DLCO, PaO 2 and PaCO 2 between and. All healthy had normal chest radiographs, were free of symptoms and not taking any medications. The study protocol was approved by the Human Ethics Review Committees of Nagasaki University School of Medicine and a signed consent form was obtained from each subject. Bronchoalveolar lavage and blood sampling With informed consent, BAL was performed as described previously 6,13,17 using a flexible fiberoptic bronchoscope (Olympus, P-2 Olympus, Tokyo, Japan). Briefly, the bronchoscope was wedged into one of the segmental or subsegmental bronchi of the right middle lobe. Then, 5 ml of sterilized saline at body temperature was instilled through the bronchoscope. The fluid was immediately

3 ENA-78 and IP1 in and 1147 retrieved by gentle suction using a sterile syringe, and the procedure was repeated three times. BALF was passed through two sheets of gauze and then centrifuged at 5g for 1 min at 4 1C. After washing twice with calcium- and magnesium-free phosphate-buffered saline (PBS, Gibco, UK), the remaining cells were suspended in PBS supplemented with 1% heat-inactivated foetal calf serum and were counted using a hemocytometer. An aliquot was then diluted to a concentration of cells=ml; and a.2 ml cell suspension was spun down onto a glass slide at 11 rpm for 2 min using a Cytospin 2 cytocentrifuge (Shandon Instruments, Sewickley, PA). The remaining fluid was centrifuged at 5g for 5 min, and supernatant was stored at 8 1C until examined. The slides were dried, fixed, and then stained using the May Giemsa method. More than 2 cells were identified using a photomicroscope. Subsets of lymphocytes in BALF were examined by direct immunofluorescence staining using fluorescein isothiocyanate-labeled murine monoclonal anti-cd3, CD4 and CD8 antibodies (Becton Dickinson, Mountain View, CA). The stained cells were analyzed on a flow cytometer (FACScan; Becton Dickinson, FACS Division). Blood for determination of ENA-78 and IP1 in serum was obtained from the patients at the same time. Serum was frozen at 8 1C until analysis. Measurement of ENA-78 and IP1 The concentrations of ENA-78 and IP1 in serum and BALF were measured by commercially available enzyme-linked immunosorbent assay kits (Quantikine; R&D systems, Minneapolis, MN) according to the protocols provided by the manufacturer. The detection limits were 15 and 1.67 pg/ml for ENA-78 and IP1, respectively. Although BAL procedures generally have a dilutional effect on the recovery of chemokines, a good correlation was observed between the original values and standardized values by albumin concentrations in BALF (data not shown). Thus, the original values of chemokine concentrations were reported in this study rather than corrected values by albumin concentrations. Statistical analysis Data was expressed as mean7standard deviation of the mean (SD). Differences between groups were examined using one-way analysis of variance. The post-hoc test used was Fisher s PLSD test. We also used Spearman s rank correlation analysis to examine the relationship between two variables. Statistical analysis was performed using StatView-J 5. software (Abacus Concepts, Berkeley, CA). A P- value o:5 denoted the presence of a statistically significant difference. Results BALF differential cell count Table 1 shows the total and differential cell counts in BALF of the three groups. Cell recovery in patients with and idiopathic was significantly higher than in normal controls (Po:1). The percentage of alveolar macrophages in patients with was significantly lower than in patients with and controls (Po:1), and those in patients with were also significantly lower than in the controls (Po:5). On the other hand, the percentage of lymphocytes in patients with was significantly higher than in patients with and controls (Po:1). No significant differences Table 1 Total and differential cell counts in BALF. (n ¼ 12) (n ¼ 1) (n ¼ 17) Total cell counts (1 5 =ml) 1.1 (.4) 3.3 (1.1) a 4.1 (1.6) a Macrophages (%) 86.9 (7.3) 55.2 (2.2) a,c 74.9 (13.1) b Lymphocytes (%) 1.8 (6.7) 39.9 (18.6) a,c 14.8 (1.7) Neutrophils (%) 1.9 (4.1) 2.8 (2.5) 5.3 (5.3) Eosinophils (%).4 (.9) 2.1 (1.7) 4.9 (8.1) Lymphocyte subsets CD3 (%) 79.8 (6.3) 83.1 (16.9) 85.7 (15.1) CD4 (%) 33.9 (11.5) 24. (9.8) c 48.2 (2.3) b CD8 (%) 45. (11.) 6.1 (18.1) b,c 35.4 (15.9) CD4/CD8.9 (.5).4 (.3) c 1.8 (1.5) Data are mean (SD). a Po.1 compared with normal subjects; b Po.5 compared with normal subjects; c Po.1 compared with.

4 1148 were observed in the percentages of neutrophils and eosinophils among the three groups. The CD4/ CD8 ratio in patients with was significantly lower than in patients with. BALF and serum concentrations of ENA-78 and IP1 As shown in Figs. 1 and 2, BALF concentrations of ENA-78 in patients with (79:5 7:5p=ml) were higher than in the control (5:9 1:pg=ml; Po:1) and patients with (25:1 13:6 pg=ml; Po:1), but the serum concentrations of ENA-78 in patients with (342:7 3298:8 pg=ml) were higher than those in the control (199:6 58:pg=ml; Po:2) and patients with (1267:1 1148:2pg=ml; Po:2). The concentrations of IP1 were elevated in BALF of patients with (27:7 122:7p=ml) compared with the control (29:5 25:pg=ml; Po:1) and patients with (87:7 99:2pg=ml; Po:1). As shown in Fig. 2, the concentrations of IP1 in serum in ENA-78 in BALF (A) IP 1 in BALF (B) P<.1 P<.1 P<.1 P<.1 Figure 1 Box plots of ENA-78 (A) and IP1 (B) concentrations in bronchoalveolar lavage fluid (BALF) of patients with idiopathic pulmonary fibrosis (), idiopathic nonspecific interstitial pneumonia (), and control subjects. ENA-78 in serum (A) IP 1 in serum (B) P<.2 P<.1 P<.2 patients with (423:3 384:p=ml) were higher than in the control (77:9 29:3pg=ml; Po:1). Correlations between chemokine concentrations and clinical data S. Nakayama et al. Figure 2 Box plots of ENA-78 (A) and IP1 (B) concentrations in serum of patients with, idiopathic, and control subjects. ENA-78 and IP1 are chemotactic factors for inflammatory cells. Accordingly, we examined in paired samples the correlation between these chemokines and BALF cell proportions. Data of all subjects (n ¼ 39) showed that the total cell number, the absolute numbers of macrophages, lymphocytes and neutrophils, and the percentages of neutrophils correlated significantly with BALF concentrations of ENA-78 (Table 2). The concentrations of IP1 also correlated significantly with the total cell number, the percentages of macrophages, lymphocytes and neutrophils, and the absolute number of lymphocytes and neutrophils in BALF of all subjects (Table 2). In patients with, the concentrations of IP1 correlated inversely with the percentages of macrophages and correlated significantly with the absolute number of lymphocytes in BALF (Table 2). In BALF obtained from patients with, the concentrations of IP1

5 ENA-78 and IP1 in and 1149 Table 2 Correlation between chemokine levels and BAL findings. Overall (n ¼ 39) (n ¼ 1) (n ¼ 17) ENA-78 IP1 ENA-78 IP1 ENA-78 IP1 Total cell counts.713 a.487 a Macrophages (%) a b (absolute number) (.715 a ) (.22) (.31) (.335) (.245) (.398) Lymphocytes (%) a a (absolute number) (.43 a ) (.93 a ) (.27) (.697 b ) (.15) (.766 a ) Neutrophils (%).431 a.327 b (absolute number) (.63 a ) (.47 b ) (.383) (.248) (.235) (.194) Data are correlation coefficient. a o.1; b o.5. correlated with the percentage and the number of lymphocytes (Table 2). There was no significant correlation between serum and BALF concentrations of ENA-78 in all subjects, while the BALF concentrations of IP1 correlated with the serum concentrations of IP1 (r ¼ :568; Po:1). BALF and serum concentrations of ENA-78 and IP1 did not correlate with any pulmonary function tests including PaO 2, VC, %VC, FEV 1, %FEV 1, or %DLCO in patients with or with the exception that there was significant correlation between serum concentrations of ENA-78 and FEV 1 in patients with (r ¼ :821; Po:5). The levels of C-reactive protein (CRP) in serum of almost all cases were within normal range and there was no significant correlation between the CRP levels and the chemokine levels in BALF and serum in the different groups of patients (data not shown). Discussion The major finding of the present study was that patients with showed elevated concentrations of ENA-78 in BALF compared with the control subjects and patients with idiopathic. Inversely, elevated IP1 BALF concentrations were observed in patients with compared with the control subjects and patients with. On the other hand, the serum concentrations of ENA-78 and IP- 1 were higher in patients than in the control, although those in patients were similar to those of the control. The concentrations of IP1 in BALF correlated with the percentage and number of lymphocytes in BALF of patients and also correlated with the absolute number of lymphocytes in BALF of patients. One of the distinct clinical features of is the better prognosis than that of. It is therefore important to clarify the differences in the pathogenesis between and idiopathic. The significant contribution of the CC chemokines such as RANTES and MCP-1 has already been reported in the pathogenesis of and idiopathic In addition, our previous study showed that the concentrations of RANTES in BALF were significantly higher in patients with compared with controls and patients with, suggesting a different profile of the CC chemokines between and. The CXC chemokines such as ENA-78 and IP1 are also major inflammatory cell attractants in and. Here, we demonstrated distinct profiles of ENA-78 and IP1 between and. The CXC subfamily can be divided into two subgroups depending on the presence or absence of the Glu-Leu-Arg (ELR) motif, which immediately precedes the first cysteine residue (ELR + CXC and ELR CXC chemokines, respectively). ENA-78 is a potent ELR + CXC chemokine and known to selectively attract and activate neutrophils. ENA-78 was initially identified from the conditioned medium of a human type II epithelial cell line A Subsequent studies showed that it is produced by other tissue cells, including monocytes and endothelial cells. 19,2 Increased ENA-78 expression has been shown in other inflammatory and infectious diseases such as rheumatoid arthritis, 21 bacterial meningitis, 22 1 and acute respiratory distress syndrome (ARDS). 23,24 In the latter condition, high concentrations of ENA-78 and IL-8 were detected in BALF and they correlated significantly with the numbers of neutrophils in BALF. 23 Keane and colleagues 15 demonstrated that lung tissues from the patients with also constitutively express more ENA-78 than control lung tissues.

6 115 Our results are in agreement with these early findings. There were no significant correlations between BALF concentrations of ENA-78 and the percentage and number of neutrophils in BALF of and patients while a significant correlation was observed in all subjects. These findings suggest that ENA-78 is not the predominant neutrophil chemoattractant in and while this study used relatively small numbers of patients. IP1 is a potent ELR CXC chemokine known to induce chemotaxis of lymphocytes, especially activated T cells. 25 IP1 is induced in monocytes/ macrophages by interferon-g: IP1 is an important mediator for the recruitment of activated lymphocytes into the lungs in certain pulmonary diseases such as sarcoidosis, 26 tuberculosis 27 and HIV-associated T cell alveolitis. 28 In sarcoidosis, IP1 is strongly expressed in sarcoid tissues and there are positive correlations between BALF concentrations of IP1, the degree of CD4 alveolitis, the expression of Th1 cytokines and CXCR3 (CXC chemokinespecific receptor) by sarcoid T cells. 26 These results suggest a role for IP1 in the mechanisms that account for sarcoid T cell activation and the development of sarcoid granuloma. In HIV patients with high intensity T cell alveolitis, lymphocytes recovered from BALF were CD8+ T cells that expressed high concentrations of CXCR3 and IFN-g; a phenotype that is characteristic of Th1 cells. These findings make IP1 an attractive candidate for inflammatory cell attraction to the lung in and idiopathic. We found elevated IP-1 BALF concentrations in patients with compared with healthy controls and patients with. On the other hand, serum concentrations of IP-1 were higher in patients than in the control. The concentrations of IP1 in BALF also correlated with the percentage and number of lymphocytes in BALF of patients and also correlated with the absolute number of lymphocytes in BALF of patients. Taken together, our results suggest that lymphocytes may play an important role in development and that IP1 contributes significantly to the accumulation of lymphocytes in the lung of various pulmonary inflammatory diseases, especially in. In the present study, the BALF concentrations of ENA-78 were elevated in patients compared with the control subjects and patients with. The elevated BALF ENA-78 concentrations may be further evidence of the importance of angiogenesis in the pathogenesis of pulmonary fibrosis. In fact, the existence of neovascularization in was originally identified by Turner Warwick, who examined the lungs of patients with widespread interstitial pneumonia, and demonstrated neovascularization leading to anastomoses between the systemic and pulmonary microvasculature and evidence of extensive vascular remodelling in areas of fibrosis. 29 The CXC chemokines have been shown to have disparate effects in mediating angiogenesis as a function of the presence or absence of ELR motif. ENA-78 and IL-8, both contain the ELR motif, which confers potent angiogenic activity, whereas members such as IP1, which lack the ELR motif, inhibit angiogenesis. Keane and colleagues demonstrated an imbalance of expression of angiogenic (ENA-78 and IL-8) vs angiostatic (IP1) factors in lung biopsy specimens from patients. 14,15 Angiogenesis plays an important role in wound healing and may contribute to the fibroproliferation and extracellular matrix deposition observed in. Keane and colleagues also showed that neutralization of macrophage inflammatory protein 2 (MIP-2), a murine functional homolog of IL-8, attenuates bleomycin-induced pulmonary fibrosis via inhibition of angiogenesis. 3 In contrast, administration of IP1 led to a reduction in pulmonary fibrosis that was mediated through the inhibition of angiogenesis. 31 These findings suggest that ENA-78 and IP1 may play important roles in the pathogenesis of through regulation of angiogenic activity. The elevated BALF levels of ENA-78 in and IP1 in, respectively, which were shown in the present study, therefore, may affect directly the difference in the prognosis of patients with these two diseases. However, to our knowledge, there are no reports of the difference in angiogenesis between and. Further studies on angiogenesis and vascular remodelling are needed to clarify whether is pathogenetically different from. In conclusion, we demonstrated in the present study the presence of elevated BALF concentrations of ENA-78 in and IP1 in. The findings suggest that ENA-78 may serve as an angiogenic factor in rather than a chemoattractant, and that IP1 may act as a predominant chemotactic factor for lymphocytes in idiopathic. Acknowledgments S. Nakayama et al. We thank A. Yokoyama for the excellent technical assistance and Dr. M. Kitaichi (Department of Laboratory Medicine and Pathology, NHO Kinki-chuo Chest Medical Center) for the valuable advice regarding pathological diagnosis. This study was supported in part by a research grant from the Ministry of Education, Science, Sports, and Culture of Japan.

7 ENA-78 and IP1 in and 1151 References 1. American Thoracic Society. Idiopathic pulmonary fibrosis: diagnosis and treatment. International consensus statement. Am J Respir Crit Care Med 2;161: Ryu JH, Colby TV, Hartman TE. Idiopathic pulmonary fibrosis: current concepts. Mayo Clin Proc 1998;73: Katzenstein AL, Myers JL. Idiopathic pulmonary fibrosis. Clinical relevance of pathologic classification. Am J Respir Crit Care Med 1998;157: Nagai S, Kitaichi M, Itoh H, Nishimura K, Izumi T, Colby TV. Idiopathic nonspecific interstitial pneumonia/fibrosis: comparison with idiopathic pulmonary fibrosis and BOOP. Eur Respir J 1998;12: Park CS, Chung SW, Ki SY, et al. Increased levels of interleukin-6 are associated with lymphocytosis in bronchoalveolar lavage fluids of idiopathic nonspecific interstitial pneumonia. Am J Respir Crit Care Med 2;162: Ishii H, Mukae H, Kadota J, et al. High serum concentrations of surfactant protein A in usual interstitial pneumonia compared with non-specific interstitial pneumonia. Thorax 23;58: Luster AD. Chemokines chemotactic cytokines that mediate inflammation. N Engl J Med 1998;338: Antoniades HN, Neville-Golden J, Galanopoulos T, Kradin RL, Valente AJ, Graves DT. Expression of monocyte chemoattractant protein 1 mrna in human idiopathic pulmonary fibrosis. Proc Natl Acad Sci USA 1992;89: Iyonaga K, Takeya M, Saita N, et al. Monocyte chemoattractant protein-1 in idiopathic pulmonary fibrosis and other interstitial lung diseases. Hum Pathol 1994;25: Car BD, Meloni F, Luisetti M, Semenzato G, Gialdroni-Grassi G, Walz A. Elevated IL-8 and MCP-1 in bronchoalveolar lavage fluids of patients with idiopathic pulmonary fibrosis and pulmonary sarcoidosis. Am J Respir Crit Care Med 1994;149: Kodama N, Yamaguchi E, Hizawa N, et al. Expression of RANTES by bronchoalveolar lavage cells in nonsmoking patients with interstitial lung diseases. Am J Respir Cell Mol Biol 1998;18: Suga M, Iyonaga K, Ichiyasu H, Saita N, Yamasaki H, Ando M. Clinical significance of MCP-1 levels in BALF and serum in patients with interstitial lung diseases. Eur Respir J 1999;14: Yoshioka S, Mukae H, Sugiyama K, et al. High-BAL fluid concentrations of RANTES in nonspecific interstitial pneumonia compared with usual interstitial pneumonia. Respir Med 24;98: Keane MP, Arenberg DA, Lynch JP, et al. The CXC chemokines, IL-8 and IP-1, regulate angiogenic activity in idiopathic pulmonary fibrosis. J Immunol 1997;159: Keane MP, Belperio JA, Burdick MD, Lynch JP, Fishbein MC, Strieter RM. ENA-78 is an important angiogenic factor in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 21;164: Travis WD, Matsui K, Moss J, Ferrans VJ. Idiopathic nonspecific interstitial pneumonia: prognostic significance of cellular and fibrosing patterns: survival comparison with usual interstitial pneumonia and desquamative interstitial pneumonia. Am J Surg Pathol 2;24: Mukae H, Iiboshi H, Nakazato M, et al. Raised plasma concentrations of a-defensins in patients with idiopathic pulmonary fibrosis. Thorax 22;57: Walz A, Burgener R, Car B, Baggiolini M, Kunkel SL, Strieter RM. Structure and neutrophil-activating properties of a novel inflammatory peptide (ENA-78) with homology to interleukin 8. J Exp Med 1991;174: Walz A, Schmutz P, Mueller C, Schnyder-Candrian S. Regulation and function of the CXC chemokine ENA-78 in monocytes and its role in disease. J Leukoc Biol 1997;62: Imaizumi T, Albertine KH, Jicha DL, McIntyre TM, Prescott SM, Zimmerman GA. Human endothelial cells synthesize ENA-78: relationship to IL-8 and to signaling of PMN adhesion. Am J Respir Cell Mol Biol 1997;17: Halloran MM, Woods JM, Strieter RM, et al. The role of an epithelial neutrophil-activating peptide-78-like protein in rat adjuvant-induced arthritis. JImmunol1999;162: Zwijnenburg PJG, de Bie HMA, Roord JJ, van der Poll T, van Furth AM. Chemotactic activity of CXCL5 in cerebrospinal fluid of children with bacterial meningitis. J Neuroimmunol 23;145: Goodman RB, Strieter RM, Martin DP, et al. Inflammatory cytokines in patients with persistence of the acute respiratory distress syndrome. Am J Respir Crit Care Med 1996;154: Keane MP, Donnelly SC, Belperio JA, et al. Imbalance in the expression of CXC chemokines correlates with bronchoalveolar lavage fluid angiogenic activity and procollagen levels in acute respiratory distress syndrome. J Immunol 22;169: Taub DD, Lloyd AR, Conlon K, et al. Recombinant human interferon-inducible protein 1 is a chemoattractant for human monocytes and T lymphocytes and promotes T cell adhesion to endothelial cells. J Exp Med 1993;177: Agostini C, Cassatella M, Zambello R, et al. Involvement of the IP-1 chemokine in sarcoid granulomatous reactions. J Immunol 1998;161: Sauty A, Dziejman M, Taha RA, et al. The T cell-specific CXC chemokines IP-1, Mig, and I-TAC are expressed by activated human bronchial epithelial cells. J Immunol 1999;162: Agostini C, Facco M, Siviero M, et al. CXC chemokines IP-1 and Mig expression and direct migration of pulmonary CD8+/ CXCR3+ T cells in the lungs of patients with HIV infection and T-cell alveolitis. Am J Respir Crit Care Med 2;162: Turner-Warwick M. Precapillary systemic-pulmonary anastomoses. Thorax 1963;18: Keane MP, Belperio JA, Moore TA, et al. Neutralization of the CXC chemokine, macrophage inflammatory protein-2, attenuates bleomycin-induced pulmonary fibrosis. J Immunol 1999;162: Keane MP, Belperio JA, Arenberg DA, et al. IFN-g-inducible protein-1 attenuates bleomycin-induced pulmonary fibrosis via inhibition of angiogenesis. J Immunol 1999;163: